Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

He had been expecting this. For weeks, ever since the new guy moved in next door, he'd been anticipating a request. The tree was a magnificent specimen, a natural border, and the new neighbor was clearly… ambitious.

"Hey, I was wondering if we could talk about that tree," the neighbor said, his hand already reaching for the trunk. He flinched when he saw my eyes, focused, unblinking.

A coldness settled in my gut, a feeling of warning. "What about it?" I asked, my voice flat.

He started to stammer and explain, but I cut him off. "I'll think about it." I turned and slowly walked back towards my house. He could see me, I was sure, from the corner of my eye.

Emotion: vigilant

Cluster: Suspicion / Vigilance
PC1 (Valence): -0.15 Negative
PC2 (Disposition): 1.11

Role in Research

This story is one of 1,000 stories generated for the emotion vigilant. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 vigilant stories, after denoising with neutral dialogue baselines, produces the vigilant emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the vigilant vector is projected through the unembedding matrix.

Promoted:
🕵0.354
🔍0.320
🔎0.320
🧐0.316
scrutin0.310
Suppressed:
L-0.499
B-0.402
H-0.372
la-0.330
S-0.318